Embodiments described herein are generally directed to improvements relating to power, latency, bandwidth and/or performance issues relating to GPU processing/caching. According to one embodiment, a system includes a producer intellectual property (IP) (e.g., a media IP), a compute core (e.g., a GPU or an AI-specific core of the GPU), a streaming buffer logically interposed between the producer IP and the compute core. The producer IP is operable to consume data from memory and output results to the streaming buffer. The compute core is operable to perform AI inference processing based on data consumed from the streaming buffer and output AI inference processing results to the memory.
Legal claims defining the scope of protection, as filed with the USPTO.
2. The method of claim 1, wherein the streaming buffer comprises a cache.
3. The method of claim 1, wherein the memory comprises dynamic random access memory.
4. The method of claim 1, wherein the memory comprises a system memory of a computer system.
5. The method of claim 1, wherein the memory comprises a dedicated graphics memory.
6. The method of claim 1, wherein said notifying the AI-specific core regarding availability of the data in the streaming buffer is performed by the streaming buffer.
7. The method of claim 1, wherein said notifying the AI-specific core regarding availability of the data in the streaming buffer is performed by the media IP.
8. The method of claim 1, wherein said notifying the media IP regarding consumption of the data by the AI-specific core is performed by the streaming buffer.
9. The method of claim 1, wherein said notifying the media IP regarding consumption of the data by the AI-specific core is performed by the AI-specific core.
10. The method of claim 1, wherein the unit of data comprises an image frame and wherein the analytical processing unit comprises a portion of the image frame.
11. The method of claim 1, wherein the media processing comprises one or more of encoding, decoding, or transcoding media to, from, or between one or more media encoding formats.
12. The method of claim 1, wherein the media analytics processing comprises performing artificial intelligence (AI) inferences.
13. The method of claim 1, wherein the AI-specific core comprises a shader core of the GPU.
15. The system of claim 14, wherein the streaming buffer comprises a local cache of the GPU.
16. The system of claim 14, wherein the memory comprises dynamic random access memory.
17. The system of claim 14, wherein system comprises a computer system and wherein the memory comprises a system memory of the computer system.
18. The system of claim 14, wherein the memory comprises a dedicated graphics memory.
19. The system of claim 14, wherein said notifying the AI-specific core regarding availability of the data in the streaming buffer is performed by the streaming buffer.
20. The system of claim 14, wherein said notifying the AI-specific core regarding availability of the data in the streaming buffer is performed by the media IP.
21. The system of claim 14, wherein said notifying the media IP regarding consumption of the data by the AI-specific core is performed by the streaming buffer.
22. The method of claim 14, wherein said notifying the media IP regarding consumption of the data by the AI-specific core is performed by the AI-specific core.
23. The system of claim 14 wherein the unit of data comprises an image frame and wherein the analytical processing unit comprises a portion of the image frame.
24. The system of claim 14, wherein the media processing comprises one or more of encoding, decoding, or transcoding media to, from, or between one or more media encoding formats.
25. The system of claim 14, wherein the media analytics processing comprises performing artificial intelligence (AI) inferences.
26. The system of claim 14, wherein the AI-specific core comprises a shader core of the GPU.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
November 11, 2020
January 2, 2024
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.